Hypothesis

Co‑administration of AAV9‑TFEB (neuronal) and AAV5‑AQP4‑enhancer (astrocytic) produces a synergistic, sleep‑dependent clearance circuit that selectively removes toxic protein aggregates while preserving essential synaptic proteins; disrupting the temporal alignment of this circuit with slow‑wave sleep abolishes the protective effect and accelerates neurodegeneration.

Rationale

Sleep‑dependent glymphatic flow relies on astrocytic AQP4 polarity, which AAV5 can modulate to increase perivascular water flux【https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0076310】. Simultaneously, neuronal TFEB overexpression driven by AAV9 boosts lysosomal autophagy, reducing intracellular Aβ and tau burden【https://doi.org/10.1523/jneurosci.0705-15.2015】. The two systems are naturally synchronized during slow‑wave sleep, when glymphatic clearance rises 80‑90%【https://gethealthspan.com/science/article/glymphatic-system-autophagy-neurodegenerative-disease-prevention】. We propose that the coupling is not merely coincidental but mechanistic: AQP4‑mediated interstitial flow convects extracellular waste toward perivascular spaces, while TFEB‑activated lysosomes in neurons degrade internalized cargo, creating a bidirectional clearance loop that optimizes proteostasis only when both are active concurrently.

Predictions

1. Mice receiving combined AAV9‑TFEB + AAV5‑AQP4‑enhancer will show a >2‑fold reduction in soluble Aβ42 and phospho‑tau levels compared with single‑vector or control groups, but only when allowed unrestricted slow‑wave sleep.
2. Selective fragmentation of slow‑wave sleep (using optogenetic arousal) will negate the benefit of dual vectors, resulting in protein aggregate levels indistinguishable from untreated controls.
3. Synaptic protein density (e.g., PSD‑95, synaptophysin) will be preserved in the dual‑vector, sleep‑intact group but depleted when sleep‑wake coupling is disrupted, indicating selective sparing of functional components.
4. Behavioral assays (Morris water maze, novel object recognition) will reveal rescued spatial memory only in the dual‑vector, sleep‑intact condition.

Experimental Design

• Groups: (1) AAV9‑TFEB + AAV5‑AQP4‑enhancer, (2) AAV9‑TFEB alone, (3) AAV5‑AQP4‑enhancer alone, (4) empty vector controls. All groups receive systemic injection at 2 months of age.
• Sleep manipulation: Sub‑cohorts undergo chronic sleep fragmentation (4 h windows of gentle handling) vs. ad libitum sleep for 8 weeks.
• Readouts: Biochemical quantification of Aβ/tau (ELISA, Western blot), immunohistochemical assessment of lysosomal activity (LAMP1, cathepsin D) and AQP4 polarity, synaptic marker density, electrophysiological slow‑wave power, and cognitive testing.
• Statistical plan: Two‑way ANOVA (vector × sleep condition) with post‑hoc Tukey; significance set at p<0.05.

Falsifiability

If dual‑vector treatment reduces aggregate load irrespective of sleep fragmentation, or if synaptic preservation occurs without intact slow‑wave waves, the hypothesis that sleep‑gated coupling is essential for synergistic clearance would be falsified. Conversely, a clear interaction where only the sleep‑intact dual‑vector group shows both biochemical and cognitive rescue would support the proposed mechanism.